Biochem. J. (2013) 453, 345–356 (Printed in Great Britain) doi:10.

1042/BJ20121885 345

Stabilization of the angiotensin-(1–7) receptor Mas through interaction with

PSD95
Weihua BIAN*1 , Licui SUN*1 , Longyan YANG*1 , Ji-Feng LI*2 , Jia HU†, Shuai ZHENG*, Ruihan GUO*, Duiping FENG‡, Qian MA*,
Xiaocui SHI*, Ying XIONG*, Xiaomei YANG*, Ran SONG*, Jianguo XU§, Songlin WANG* and Junqi HE*†3
*Department of Biochemistry and Molecular Biology, Capital Medical University, Beijing 100069, China, †Proteomic Research Center, Capital Medical University, Beijing 100069,
China, ‡Department of Radiology, First Hospital of Shanxi Medical University, Taiyuan 030001, China, §Shaoxing Second Hospital, Zhejiang 312000, China, and Molecular
Laboratory for Gene Therapy and Tooth Regeneration, Capital Medical University School of Stomatology, Beijing 100050, China

The functions and signalling mechanisms of the Ang-(1–7)
[angiotensin-(1–7)] receptor Mas have been studied extensively.
However, less attention has been paid to the intracellular
regulation of Mas protein. In the present study, PSD95
(postsynaptic density 95), a novel binding protein of Mas
receptor, was identified, and their association was characterized
further. Mas specifically interacts with PDZ1-2, but not the
PDZ3, domain of PSD95 via Mas-CT (Mas C-terminus), and
the last four amino acids [ETVV (Glu-Thr-Val-Val)] of Mas-CT
were determined to be essential for this interaction, as shown
by GST pull-down, co-immunoprecipitation and confocal co-
localization experiments. Gain-of-function and loss-of-function
studies indicated that PSD95 enhanced Mas protein expression
by increasing the stabilization of the receptor. Mas degradation
was robustly inhibited by the proteasome inhibitor MG132 in
time- and dose-dependent manners, and the expression of PSD95
impaired Mas ubiquitination, indicating that the PSD95–Mas
association inhibits Mas receptor degradation via the ubiquitin–
proteasome proteolytic pathway. These findings reveal a novel
mechanism of Mas receptor regulation by which its expression
is modulated at the post-translational level by ubiquitination, and
clarify the role of PSD95, which binds directly to Mas, blocking
the ubiquitination and subsequent degradation of the receptor
via the ubiquitin–proteasome proteolytic pathway.
Key words: angiotensin-(1–7), G-protein-coupled receptor
(GPCR), Mas, PDZ domain, postsynaptic density 95 (PSD95),
protein interaction, ubiquitination.

INTRODUCTION
The renin–angiotensin system is one of the most comprehensively
researched and clinically relevant homoeostatic systems in human
physiology. As an important bioactive peptide of the renin–
angiotensin system, Ang-(1–7) [angiotensin-(1–7)] is involved
in many biological processes such as neural plasticity, memory
and anxiety, and possesses anti-angiogenesis, vasodilatation, anti-
proliferation, anti-fibrosis, anti-hypertrophy and anti-thrombosis
properties via a direct interaction with its special receptor Mas
[1–5].
In the brain, the Mas receptor is critical for certain properties of
the central nervous system, including neural plasticity, memory
and anxiety. Studies have shown that brief seizure episodes lead
to a significant and transient increase in Mas mRNA expression
in the hippocampus using in situ hybridization and RNase
protection assays [6]. The genetic deletion of Mas abolishes the
enhancement of LTP (long-term potentiation) induced by Ang-(1–
7) in the CA1 region of the hippocampus, implicating its important
function in learning and memory mechanisms [7]. Alterations of
spatial learning and anxiety behaviours in Mas-deficient mice are
demonstrated by their performance in the elevated plus maze task
and the Morris water maze [8]. In addition, the regulation of the
brain Mas receptor has a direct impact on cardiovascular diseases
including hypertension and heart disease. In sinoaortic denervated
rats, the release of hypothalamic Ang-(1–7) can activate the Mas
receptor and cause a reduction in blood pressure [9].
The critical role of Ang-(1–7) in many signalling pathways
through its association with the Mas receptor has been reported
in several studies. Genetic ablation of Mas abolishes Ang-(1–
7)-mediated vasodilation of the aorta, resulting in impaired heart
function [10,11]. Activation of the Mas receptor by Ang-(1–7)
stimulates nitric oxide release from endothelial cells, which
involves the PI3K (phosphoinositide 3-kinase)–Akt axis, causing
vasodilation, inhibition of cell growth and blocking AT1R [AngII
(angiotensin II) type 1 receptor]-mediated vasoconstriction
[12,13]. In mouse bone-marrow-derived dendritic cells, Ang-
(1–7) enhances AngII-induced phosphorylation of ERK1/2
(extracellular-signal-regulated kinase 1/2) [14]. Ang-(1–7)
also inhibits AngII-induced smooth muscle cell proliferation
and migration through negative modulation of AngII-induced
ERK1/2 activation [15]. Rabelo et al. [13] reported that Ang-
(1–7) acts through the receptor Mas to counteract AngII- and
AT1R-dependent c-Src activation, ERK1/2 phosphorylation
and the generation of ROS (reactive oxygen species) in human
endothelial cells [13].
Most studies on the Mas receptor have focused on its molecular
mechanisms and physiological functions, but much less attention
has been paid to the intracellular regulation of Mas receptor
signalling and its post-translational modification. The Mas gene

Abbreviations used: Ang-(1–7), angiotensin-(1–7); AngII, angiotensin II; β1 AR, β1 -adrenergic receptor; AT1R, AngII type 1 receptor; BHK, baby-hamster
kidney; CHX, cycloheximide; CT, C-terminus; DMEM, Dulbecco’s modified Eagle’s medium; ERK1/2, extracellular-signal-regulated kinase 1/2; GAPDH,
glyceraldehyde-3-phosphate dehydrogenase; GPCR, G-protein-coupled receptor; HRP, horseradish peroxidase; LTP, long-term potentiation; nNOS,
neuronal nitric oxide synthase; PI3K, phosphoinositide 3-kinase; PSD95, postsynaptic density 95; PTEN, phosphatase and tensin homologue deleted on
chromosome 10; qPCR, quantitative real-time PCR; SHR, spontaneously hypertensive rat; WKY, Wistar Kyoto; wt, wild-type.
1
These authors contributed equally to this work.
2
Present address: Department of Respiratory Medicine, Beijing Chao-Yang Hospital, Capital Medical University, Beijing 100020, China.
3
To whom correspondence should be addressed (email jq_he@ccmu.edu.cn).


c The Authors Journal compilation 
c 2013 Biochemical Society
346 W. Bian and others
encodes a protein belonging to the classic family of GPCRs
(G-protein-coupled receptors) characterized by seven transmem-
brane domains [16]. The activation of GPCRs is generally
regulated by post-translational modifications and/or association
of the receptor with its binding molecules [17]. Ubiquitination,
acylation and phosphorylation are some of the most common post-
translational modifications involved in GPCR overall expression,
subcellular trafficking and signalling. However, there are currently
no reports on the post-translational modification of the Mas
receptor or the intracellular regulation of Mas signalling.
Investigating the Mas receptor-binding molecules and its post-
translational modifications are therefore of great interest.
The Mas receptor contains a conserved typical PDZ domain-
binding motif [ETVV (Glu-Thr-Val-Val)] at its CT (C-terminus).
PDZ domains are modular protein-interaction domains and play
important roles in receptor signalling and functions [18,19]. In
the present study, we identified PSD95 (postsynaptic density 95)
associated with the Mas receptor, and the stabilization of the Mas
receptor is regulated by PSD95 via direct interaction and possibly
through the inhibition of Mas ubiquitination.
MATERIALS AND METHODS
Preparation of plasmids
The pET30A vectors encoding individual PDZ domains of
PSD95, pCGN-HA-Mas and pGW-myc-PSD95 constructs were
kindly provided by Dr Randy Hall (Emory University, Atlanta,
GA, U.S.A.). The wt (wild-type) and its mutant (V325A) of
human Mas were amplified by PCR from the pCGN-HA-Mas
construct and subcloned into pcDNA3-FLAG and pEGFP-C1
expression vectors respectively. The CT of Mas (the last 27 amino
acids of Mas, Mas-CT) wt and its various mutants were amplified
by PCR from the pCGN-HA-Mas construct and subcloned into
pGEX-4T-1. All of the DNA constructs were individually verified
by DNA sequencing.
Western blotting and antibodies
Samples were run on SDS/PAGE and then transferred on to
a nitrocellulose membrane. The membrane was blocked in
TBST [TBS (20 mM Tris/HCl, pH 7.5, and 500 mM NaCl) with
0.05 % Tween 20] containing 5 % (w/v) non-fat dried milk for
1 h at room temperature (23–25 ◦ C). Proteins of interest were
probed with corresponding primary antibodies followed by HRP
(horseradish peroxidase)-conjugated anti-rabbit or anti-mouse
secondary antibodies. Immunoreactive bands were visualized by
ECL detection reagents (Applygen Technologies) and analysed
with NIH Image 1.62.
The anti-His6 and anti-GFP antibodies were obtained from
MBL. Anti-FLAG antibody was from Applygen Technologies.
Anti-PSD95 antibody was obtained from Sigma–Aldrich. Anti-
Mas antibody was from Novus Biologicals. HRP-conjugated anti-
GST antibody was obtained from GE Healthcare. Anti-ubiquitin
was from Santa Cruz Biotechnology. Anti-GAPDH (glycera-
ldehyde-3-phosphate dehydrogenase), anti-actin and HRP-
conjugated secondary antibodies were obtained from ZSGB-BIO.
Cells and transfection
COS-7, BHK (baby-hamster kidney), LN229 and C6 glioma
cells were maintained in DMEM (Dulbecco’s modified Eagle’s
medium) plus 10 % (v/v) FBS and 1 % penicillin/streptomycin
(Sigma–Aldrich). Primary cortical neurons were prepared from
newborn (under 24 h old) Wistar rats following the method
described in [20]. All animal experiments conformed to the

c The Authors Journal compilation 
c 2013 Biochemical Society
institutional and national ethical guidelines for the care and
use of animals. The cells were cultured on polylysine-coated
glass coverslips in DMEM supplemented with 10 % (v/v)
FBS, 10 % (v/v) horse serum and 1 % penicillin/streptomycin.
Cell transfection was performed with LipofectamineTM 2000
(Invitrogen). For cell stable transfection, constructs of GFP, GFP–
Mas wt or GFP–Mas-V325A were transfected into C6 cells
respectively with LipofectamineTM 2000, and selected with the
growth medium containing 1200 μg/ml neomycin (Amresco).
GST-fusion protein pull-down assay
GST-fusion proteins were purified from bacteria with glutathione–
Sepharose 4B beads (Sigma–Aldrich) according to the
manufacturer’s protocol and resuspended in harvest buffer
(50 mM NaCl, 10 mM Hepes, pH 7.4, 5 mM EDTA, 1 mM
benzamidine and 0.1 % Tween 20). Equal amounts of bead-
conjugated GST-fusion proteins were incubated with 20 mg of
rabbit brain lysates or equal amounts of His6 -tagged PDZ domains
of PSD95 protein. After incubation with gentle rotation at 4 ◦ C for
3 h, the beads were washed extensively four times with washing
buffer (100 mM NaCl, 10 mM Hepes, pH 7.4, 5 mM EDTA,
1 mM benzamidine, 3 % BSA and 0.1 % Tween 20) and once
with washing buffer without BSA. Proteins were eluted from the
beads by boiling in sample buffer (50 mM Tris/HCl, 100 mM
DTT, 2 % SDS, 0.1 % Bromophenol Blue and 10 % glycerol),
resolved on SDS/PAGE and analysed by Western blotting.
MALDI–TOF/TOF-MS analysis
Samples from GST pull-down assays were run on SDS/PAGE.
Protein bands of interest were excised from Coomassie Blue-
stained gels, cut into small pieces, washed three times
with washing buffer (25 mM ammonium bicarbonate in 50 %
acetonitrile, pH 8.0) for 15 min each time. The samples were then
reduced and alkylated with DTT (10 mM) and iodoacetamide
(55 mM) subsequently. The gel pieces were then subjected to in-
gel digestion, as described previously [21]. The proteins were then
identified by combining peptide mass fingerprints and sequence
tags obtained by MALDI–TOF/TOF-MS on an ultrafleXtremeTM
MS instrument from Bruker Daltonics. For analysis, 0.5 μl of
the digested sample was spotted on to an anchorchip target
(Bruker Daltonics). The droplet of the sample was dried at room
temperature, and overlaid with 0.5 μl of matrix solution [5 mg/ml
L-cyano-4-hydroxycinnamic acid in 50 % acetonitrile and 0.1 %
TFA (trifluoroacetic acid)]. The mass spectra were calibrated
externally using the monoisotopic [M + H] + ion of peptide
calibration standards (Bruker Daltonics). Mass data acquisitions
were piloted by flexControl software (version 3.3). The positive
MALDI–TOF spectra and MS/MS LIFT spectra of five selected
ions were collected, and peak-lists were generated using SNAP
peak detection algorithm, signal/noise threshold higher than 6
and TopHat baseline subtraction in FlexAnalysis 3.3 (Bruker
Daltonics). The MS spectra and MS/MS spectra were combined
by Biotools (Bruker Daltonics), then subjected to the Mascot
search engine for protein identification. The mass spectra data
were searched in mammalian sequences in the database of NCBInr
20110419, using the following parameters: monoisotopic; peptide
mass tolerance, +−100 p.p.m.; fragment mass tolerance, + −0.5 Da;
missed cleavages, 1.
Blot overlay assay
A blot overlay assay was performed as described previously [22].
Briefly, His6 -tagged PDZ domains of PSD95-fusion protein were

purified, run on SDS/PAGE (10 % gels) and then transferred on
to a nitrocellulose membrane (Millipore). The membrane was
blocked and incubated with different concentrations of GST–
Mas-CT fusion proteins overnight at 4 ◦ C. The blots were washed
three times and then incubated with HRP-conjugated anti-GST
antibody for 1 h at room temperature. GST–Mas-CT fusion
protein overlaid on His6 –PDZ fusion proteins were visualized
via ECL and analysed with NIH Image 1.62.
Co-immunoprecipitation
Co-immunoprecipitation was performed as described previously
[22]. Briefly, transfected cells were harvested or rabbit brain
was homogenized in ice-cold lysis buffer. Supernatants were
incubated with anti-FLAG beads (for transfected cells) or anti-
PSD95 antibody (for brain lysates) pre-bound to Protein A/G–
agarose beads (Calbiochem). The immunoprecipitated proteins
were then analysed by Western blotting.
Immunofluorescence co-localization
Immunofluorescence co-localization was performed as described
previously [22]. Briefly, BHK cells were transfected with
GFP–Mas and/or PSD95 constructs. After fixation and
permeabilization, cells were stained with anti-PSD95 antibody
and anti-(proteasome 20S α/β) antibody (Abcam) followed
by rhodamine-conjugated anti-(mouse IgG) antibody (ZSGB-
BIO), Alexa Fluor® 647-conjugated anti-(rabbit IgG) antibody
(Invitrogen) and Hoechst 33258 (5 μg/ml). The cellular
distribution of these proteins was then visualized under a confocal
microscope (Leica Microsystems, LAS AF-TCS SP5).
PSD95 siRNA knockdown assay
siRNA duplexes directed against PSD95 (siRNA1, 5 -GUUCCA-
UCGUUCGCCUCUA-3 , and siRNA2, 5 -GAUAUGAGUUGC-
AGGUGAA-3 ), and the scrambled control (5 -UUCUCCGA-
ACGUGUCACGU-3 ) were synthesized by Sigma–Aldrich. To
perform the siRNA experiments, human glioblastoma LN229
cells with endogenous PSD95 and Mas were grown to 80 %
confluence in 35-mm-diameter dishes, whereas the neurons were
primarily cultured in vitro for 5–7 days to 60–80 % confluence.
Then, 60 nM PSD95 siRNA (mixture of equal amounts of
PSD95 siRNA1 and siRNA2) were transfected into cells with
2 μl of LipofectamineTM 2000 according to the manufacturer’s
instructions. The cells were then harvested, and protein expression
was analysed after 48 h of transfection.
Total RNA extraction and qPCR (quantitative real-time PCR)
Total cellular RNA was extracted with TRIzol® reagent. The
primer sequences for PSD95 were forward, 5 -ACTGCATCCTT-
GCGAAGCAAC-3 , and reverse, 5 -CGTCAATGACATGAAG-
CACATCC-3 . The primer sequences for Mas were forward,
5 -AACTGGCAGGAACGCCTCA-3 , and reverse, 5 -CTCATC-
CGGAAGCACAGGAAC-3 . The Mas and PSD95 mRNA levels
were determined using a Brilliant II SYBR® Green qPCR Master
Mix kits respectively (Agilent Technologies), and the samples
were normalized with an internal control of GAPDH.
RESULTS
PSD95 was identified as a novel Mas-binding protein
Although the Mas protein is distributed widely in various tissues,
it was first cloned from and highly expressed in the brain [16].
Interaction of PSD95 and Mas receptor 347
We therefore used rabbit brain lysates to screen Mas-binding
proteins in the present study. Mas-CT wt and Mas-V325A mutant
GST-fusion proteins were purified and subsequently used in GST
pull-down experiments. Equal amounts of bead-conjugated GST-
fusion proteins were incubated with rabbit brain tissue lysates.
The GST moiety of the GST–Mas-CT fusion protein bound non-
specifically to multiple bands from brain lysates, whereas Mas-
CT wt was associated with several additional interacting bands
(Figure 1A). Interestingly, among these binding proteins, the band
indicated with an asterisk (*) was detected with a very weak signal
in pull-down complexes of Mas-CT-V325A (Figure 1A). The
Mas receptor contains a conserved PDZ domain-binding motif
(ETVV) at its CT, and mutation of the last valine residue to alanine
(V325A) theoretically abolishes the interaction of Mas with PDZ
domain proteins. Thus the band indicated with an asterisk was
analysed further by MS and was identified as PDZ protein of
PSD95 (Figure 1B). To verify the interaction of the Mas-CT with
PSD95, GST–Mas-CT was used for pull-down assays from rabbit
brain lysates, and protein complexes were probed with an anti-
PSD95 antibody. As shown in Figure 1(C), the robust PSD95
protein appeared in the GST–Mas-CT wt pull-down complex
and no signal was detected in the Mas-CT-V325A pull-down
complex, indicating that GST–Mas-CT wt associated strongly
with PSD95, and mutation of the terminal valine residue of Mas
to alanine (Mas-V325A) completely abolished this interaction.
These findings confirmed the interaction of the Mas-CT with
PSD95, and suggested the critical importance of the CT of Mas
in mediating this interaction.
Interaction of Mas-CT with PSD95 PDZ1-2
As PDZ1-2 of PSD95 possesses a binding preference that is quite
different from that of PDZ3, individual His6 –PDZ1-2 or His6 –
PDZ3 domain of PSD95-fusion proteins was subjected to pull
down by GST–Mas-CT to explore the structural determinants of
the Mas–PSD95 interaction (Figure 2A). A strong PSD95 PDZ1-
2 signal was detected in the GST–Mas-CT pull-down complex,
and conversely only a very weak signal for His6 –PDZ3 of PSD95
was detected, suggesting that PDZ1 and PDZ2 were the principal
domains mediating the PSD95–Mas interaction.
As the ETVV motif of the Mas-CT is a canonical class I PDZ
domain-binding motif, we explored further the Mas-CT–PSD95
association by generating a series of mutations of Mas-CT with
alanine substitutions in this motif. Mas-CT wt (ETVV) or its
various mutants (ATVV, EAVV, ETAV and ETVA) was incubated
with His6 –PDZ1-2 of PSD95, and the precipitation complexes
were detected with an anti-His6 antibody (Figure 2B). ETVV
was strongly associated with His6 –PDZ1-2 of PSD95, whereas
the mutation of any of the terminal four amino acid residues to
alanine almost completely impaired this interaction.
In addition to the fusion protein pull-down experiments, we
used the reverse experiments to assess the binding ability of the
Mas–PSD95 complex via a blot overlay assay using different
concentrations of GST–Mas-CT proteins (5–800 nM). The results
were plotted in Figure 2(C), which estimated the binding affinity
of PSD95 PDZ1-2 to the Mas-CT as 40 nM, revealing the
relatively high affinity of this interaction. Consistent with the GST
pull-down results, PSD95 PDZ3 did not detectably associate with
Mas-CT according to the overlay assay (results not shown).
PSD95 interacts with Mas in cells and native tissues
The interaction between full-length Mas and full-length PSD95
within cells was examined using co-immunoprecipitation assays.

c The Authors Journal compilation 
c 2013 Biochemical Society
348 W. Bian and others

Figure 1 Proteomic screen defines the Mas-binding partners and identifies PSD95 as a novel Mas-associated protein
(A) Screening of Mas-CT-binding partners by GST pull-down assay. Purified GST–Mas-CT fusion proteins were adsorbed to glutathione–agarose beads, and used to pull down lysates of rabbit
brain. The precipitates were run on SDS/PAGE gels and then visualized by Coomassie Blue staining. Positions of molecular-mass markers are shown on the left in kDa. (B) PSD95 was analysed as a
novel Mas-associated protein by MS. The protein band indicated in (A) with an asterisk (*), which strongly bound to the Mas-CT wt was excised from the Coomassie Blue-stained gel and analysed
by MS. Panel a, peptide mass fingerprinting of the protein band indicated in (A) with the asterisk generated by tryptic digestion. Panel b, MS/MS spectra of the peptide ion with m /z 1618.968, which
selected from MS spectra of the protein band indicated in (A) with the asterisk. MS/MS analysis suggested that this protein band was PSD95. Intensity (intens.) is measured in arbitrary units (a.u.).
(C) PSD95 was identified as Mas-CT-associated protein by Western blotting. Equal amounts of purified GST-fusion protein beads of Mas-CT wt or its V325A mutant were used respectively for pull
down of rabbit brain lysates, and the pull-down complex was subjected to Western blotting with an anti-PSD95 antibody (upper panel). Coomassie Blue staining revealed equal loading of the fusion
proteins (lower panel). Molecular masses are indicated in kDa.

The expression levels of FLAG–Mas wt and FLAG–Mas-V325A
were adjusted to similar levels through transfection with different
amounts of the respective constructs. As expected, PSD95 was
strongly co-immunoprecipitated with FLAG–Mas wt, whereas
minimal amounts were detected in the FLAG-Mas-V325A
immunoprecipitation complexes, indicating that full-length Mas
associates with PSD95 in cells, and again revealing the importance
of the terminal valine residue for their interaction (Figure 3A).
We examined further the interaction of Mas with PSD95 in
the absence and presence of agonist stimulation. As a result,
PSD95 was robustly co-immunoprecipitated with FLAG–Mas,
but treatment with Ang-(1–7), an agonist of Mas receptor, had no
detectable effect on the interaction of Mas receptor with PSD95
protein (Figure 3B). To examine the interaction of endogenous
Mas and PSD95 in rabbit brain tissue, solubilized brain lysates
were incubated with anti-PSD95 antibody linked to Protein A/G–
agarose beads. A robust Mas signal was detected in PSD95 co-
immunoprecipitation complexes, but no signal was detected in the
control samples (Figure 3C), which indicated the existence of a
physical complex between Mas and PSD95 in native tissues.
The potential co-localization of Mas and PSD95 in cultured
cells was examined via immunofluorescence microscopy.
BHK cells were used during these experiments in order to
better visualize cell morphology. BHK cells were co-transfected
with constructs of PSD95 as well as GFP–Mas wt or GFP–
Mas-V325A. In cells transfected with these constructs alone,
PSD95 was located exclusively in the plasma membrane, forming
a smooth rim around the cell [23,24] (Figure 4B, panel b).
When GFP-tagged Mas was expressed alone, it was observed
predominantly in the cytoplasm [25] (Figure 4A, panel a), partly
overlapping with the proteasome 20S α/β subunits (Figure 4A,
panel e). When Mas wt was co-expressed with PSD95, a
significant fraction of total Mas wt was retained in the plasma
membrane (Figure 4C, panel a), where it exhibited excellent
co-localization with PDS95 (Figure 4C, panel d). Conversely,
when the Mas mutant was co-expressed with PSD95, the
mutant receptors remained almost exclusively distributed in
the cytoplasm (Figure 4D, panel a), exhibiting little overlap
with PSD95 immunolabelling (Figure 4D, panel d). However,
a remarkable co-localization of Mas-V325A and proteasome 20S
α/β subunits was observed in proteasome (Figure 4D, panel e)
compared with no co-localization of Mas wt and proteasome 20S
α/β subunits (Figure 4C, panel e). These data were in agreement
with the findings of co-immunoprecipitation experiments and
confirmed the interaction between Mas and PSD95 in cells.
These data further indicated that PSD95–Mas association


c The Authors Journal compilation 
c 2013 Biochemical Society
 Interaction of PSD95 and Mas receptor 349

Figure 2 High-affinity interaction between PSD95 and Mas-CT via the ETVV motif of the Mas and the PDZ1-2 domains of PSD95
(A) Mas-CT specifically binds to the PDZ1-2 domains of PSD95. GST–Mas-CT beads were used to pull down His6 –PSD95 PDZ1-2- or PDZ3-fusion proteins. Precipitates were subjected to Western
blotting with an anti-His6 antibody. A robust signal corresponding to the PDZ1-2 domains of PSD95 was detected in the GST–Mas-CT pull-down complexes (top panel). Western blotting and
Coomassie Blue staining showed equal loading of the fusion proteins respectively (middle and bottom panels). (B) Mas-CT selectively associates with the PDZ1-2 domains of PSD95. Purified
GST-fusion proteins corresponding to either Mas-CT wt (denoted by its last four amino acids, ETVV) or point-mutated versions of the wt (denoted by sequential replacement of each of the last four
amino acids with alanine) were used to pull down the His6 -tagged PSD95 PDZ1-2 domains from the bacterial lysates. Pull-down precipitates were subjected to Western blotting with an anti-His6
antibody. Coomassie Blue staining confirmed that equivalent amounts of various GST-fusion proteins were present in each sample. (C) The overlay assay results show a relatively high affinity
between Mas-CT and PSD95 PDZ1-2. Nitrocellulose membrane containing 1 μg of His6 /S-tagged PSD95 PDZ1-2 was incubated with GST–Mas-CT at different concentrations ranging from 5 to
800 nM. PSD95 PDZ1-2-specific binding did not increase significantly between 400 and 800 nM, therefore the binding observed at 800 nM was defined as the ‘maximal’ binding. The amount of
binding at the other concentrations was expressed as a percentage of the maximal binding within each experiment. Results are means +− S.D. (n = 3). Molecular masses are indicated in kDa in (A)
and (B). IB, immunoblot.

inhibits subcellular localization of Mas with proteasome in the
cytoplasm.
PSD95 enhances Mas protein stabilization
The functional significance of the Mas–PSD95 interaction was
explored next. We observed that the expression level of Mas
was significantly increased when it was co-expressed with
PSD95 compared with Mas expressed alone (equal doses of Mas
constructs were used in the transfection) (Figure 5A), suggesting
that the Mas–PSD95 interaction may play a role in the regulation
of Mas expression levels. To examine whether the effect of PSD95
on Mas expression was dependent on the Mas–PSD95 interaction,
COS-7 cells were transfected with Mas wt or Mas-V325A in
the presence or absence of PSD95, and Mas protein levels
were determined by Western blotting. As shown in Figure 5(B),
PSD95 overexpression significantly increased the protein level
of Mas wt, but not that of the Mas-V325A mutant, indicating
that disruption of the Mas–PSD95 interaction suppressed the Mas
protein expression up-regulated by PSD95.
To verify the results of the PSD95-overexpression experiments,
endogenous PSD95 expression was knocked down in C6 (results
not shown) and LN229 (Figure 5C) cells. RNAi-mediated
knockdown of PSD95 caused a significant reduction in the PSD95
expression of approximately 70 % in LN229 cells. Meanwhile,
the protein expression of Mas was co-ordinately decreased
by approximately 30 % of the scrambled control (P < 0.05)
(Figure 5C). These results confirmed that PSD95 increases the
expression of Mas. To test further whether endogenous Mas
protein could also be modulated by PSD95 expression in primary
cultured cells, PSD95 was knocked down in primary neuron
cells with PSD95-specific siRNA, which resulted in a significant
reduction in PSD95 expression of approximately 30 % and
a concomitant decrease in endogenous Mas of approximately
30 % (Figure 5D). Taken together, the results of our PSD95-
overexpression and -knockdown experiments demonstrated that
the regulation of Mas protein expression is mediated by its
interaction with PSD95.
To explore the mechanism of Mas expression regulated by
PSD95, the Mas mRNA levels were examined next. As shown in
Figure 6(A), Mas mRNA levels did not decrease after knocking
down PSD95 expression in human glioblastoma LN229 cells,
indicating that the regulation of Mas protein expression by PSD95
is independent of Mas mRNA transcription, and may involve
the post-translational modification of Mas based on its direct
interaction with PSD95.
To test this hypothesis, we measured the turnover rate of the
Mas protein upon binding to PSD95. COS-7 cells transfected with
Mas constructs were treated with different doses of the protein
synthesis inhibitor CHX (cycloheximide) for 12 h. Figure 6(B)
shows the dose-dependent decrease in Mas protein levels after
exposure to CHX (from 0 to 50 ng/μl). Because no significant
change in Mas protein expression was detected in response to


c The Authors Journal compilation 
c 2013 Biochemical Society
350 W. Bian and others

Figure 3 Association of full-length Mas with PSD95 in cells

(A) Cellular association of full-length FLAG–Mas with Myc–PSD95. COS-7 cells were transfected with Myc–PSD95 and either FLAG–Mas wt or mutant FLAG–Mas-V325A constructs. Cell lysates
were incubated with the anti-FLAG antibody coupled to beads to immunoprecipitate FLAG-tagged receptors, and the precipitated complexes were analysed by Western blotting with an anti-PSD95
antibody to examine the interaction of Mas with PSD95 (top panels). All lysates were probed with anti-FLAG or anti-PSD95 antibody to visualize the relative equal levels of the receptors or
PSD95 (bottom panels). (B) Cellular interaction of Mas receptor with PSD95 in the absence and presence of agonist stimulation. COS-7 cells were transfected with FLAG–Mas wt in the presence or
absence of Myc–PSD95, and the cells were harvested after Ang-(1–7) (0.1 μM) treatment for different times (0, 2, 15 and 60 min). Cell lysates were incubated with the anti-FLAG antibody coupled to
beads to immunoprecipitate FLAG-tagged receptors, and the precipitated complexes were analysed by Western blotting with an anti-PSD95 antibody to examine the interaction of Mas with PSD95.
(C) Mas and PSD95 interact in native tissues. Solubilized lysates from homogenized rabbit brain tissues were subjected to immunoprecipitation with or without the anti-PSD95 antibody. The lysates
were probed with anti-Mas or anti-PSD95 antibodies to confirm the presence of Mas and PSD95 respectively. Co-immunoprecipitated Mas was then probed with the anti-Mas antibody by Western
blotting. Immunoprecipitation of the lysates with the anti-PSD95 irrelevant antibody was used as the negative control. Blots are representative of three to five independent experiments. Molecular
masses are indicated in kDa. IB, immunoblot; IP, immunoprecipitation.

CHX doses between 25 and 50 ng/μl, the 25 ng/μl CHX dose
was used for subsequent experiments with COS-7 cells transiently
transfected with various combinations of constructs including
Mas wt, Mas-V325A, Mas wt–PSD95 or Mas-V325A–PSD95.
The cells were harvested at different time points as indicated
in Figure 6(C), and the protein levels of Mas and PSD95 were
assessed by Western blotting. The protein expression levels
of Mas significantly decreased in a time-dependent manner when
Mas was expressed alone. When PSD95 was overexpressed,
however, the Mas wt protein turnover rate was significantly
reduced (P < 0.05), and Mas wt protein levels were significantly
higher than that of similarly treated Mas-V325A. These data
indicated that PSD95-mediated up-regulation of Mas protein
expression was independent of Mas mRNA transcription, and Mas
protein stabilization was enhanced by interaction with PSD95.
PSD95 enhances Mas expression by attenuation of the
ubiquitin–proteasome proteolytic Mas protein
The effect of PSD95 on the stabilization of the Mas protein was
probably mediated by the inhibition of Mas protein degradation.
Intracellular protein degradation relies mainly on the lysosomal
degradation and ubiquitin–proteasome pathways. To explore the
mechanism of Mas protein expression up-regulated by PSD95,
C6 cells stably transfected with Mas constructs were treated
with chloroquine (100 μM), a lysosome inhibitor, or MG132


c The Authors Journal compilation 
c 2013 Biochemical Society
 Interaction of PSD95 and Mas receptor 351

Figure 4 Co-localization of Mas and PSD95 in BHK cells

BHK cells were transiently transfected with GFP–Mas wt alone (A), GFP–Mas-V325A (not shown), Myc–PSD95 alone (B), GFP–Mas wt and Myc–PSD95 (C), or GFP–Mas-V325A and Myc–PSD95
(D). After fixation and permeabilization, cells were stained with an anti-PSD95 antibody and anti-(proteasome 20S α/β) antibody followed by rhodamine-conjugated anti-(mouse IgG) antibody and
Alexa Fluor® 647-conjugated anti-(rabbit IgG) antibody. Nuclei were stained with Hoechst 33258. Co-localization of Mas with PSD95 is shown in yellow following merging of the two individual
images, as shown in panels d. Co-localization of Mas with the proteasome 20S α/β subunit is shown in brown following merging of the two individual images, as shown in panels e. Images are
representative of three independent experiments.

(10 μM), a proteasome inhibitor. As shown in Figure 7(A),
a marked increase in Mas protein expression was detected in
response to MG132 treatment, whereas no detectable increase
in Mas protein expression was observed in chloroquine-treated
cells, suggesting that the degradation of the Mas protein occurs via
the proteasome pathway. To confirm further the involvement of the
proteasome degradation pathway, C6 cells stably transfected with
Mas constructs were treated with different doses of MG132. As
shown in Figure 7(B), MG132 dose-dependently elevated Mas
protein expression, which reached saturation levels with 10 μM
MG132. We also detected degradation of endogenous Mas protein
in C6 cells, in which Mas receptor moderately expressed. C6 cells
were treated with MG132 for different times (0, 3, 9 and 12 h). As
shown in Figure 7(C), endogenous Mas protein levels significantly
increased time-dependently. These data indicated further that the
degradation of the Mas protein occurs via the proteasome pathway.
To verify further that the PSD95 regulation of Mas stabilization
is mediated by the ubiquitin–proteasome pathway, we assessed
the effect of PSD95 expression on Mas protein ubiquitination in
COS-7 cells. FLAG-Mas wt or its V325A mutant with the same
dose of Mas constructs was transfected into COS-7 cells in the
presence or absence of PSD95. The cells were treated with MG132
for 12 h to inhibit the degradation of ubiquitin-conjugated Mas
and facilitate the detection of the ubiquitinated Mas molecule.
The similar levels of FLAG–Mas proteins were exhibited in all
lanes after Western blotting (Figure 7D), as PSD95 has little or no
effect on total Mas protein expression levels when cells were pre-
treated with MG132. The Mas protein was then immunoprecipit-
ated with the anti-FLAG antibody, and the ubiquitination of Mas
was analysed by immunoblotting with an anti-ubiquitin antibody.
As shown in Figure 7(D), a remarkable amount of ubiquitinated
Mas wt protein was detected in cells that were not transfected
with PSD95, whereas PSD95 co-expression reduced Mas wt
ubiquitination (P < 0.05). In contrast, the ubiquitination of the
Mas-V325A mutant was not affected by PSD95 co-expression in
comparison with that of mutant Mas expression alone, suggesting
that Mas ubiquitination was inhibited by the interaction with
PSD95. These data were consistent with the results from our
confocal microscopy data that Mas partly co-localized with
proteasomes in the cytoplasm, and the receptor was anchored
in the plasma membrane by PSD95 co-expression, inhibiting Mas
co-localization with the proteasome. Taken together, these data
indicate that PSD95 may inhibit Mas receptor ubiquitination, thus
suppressing the proteasome pathway-mediated degradation of the
receptor and enhance Mas protein stabilization.
DISCUSSION
The functions and signalling mechanisms of the Mas receptor
have been studied extensively. However, much less attention
has been paid to the intracellular regulation of Mas protein. In
the present study, we screened Mas-binding proteins by GST


c The Authors Journal compilation 
c 2013 Biochemical Society
352 W. Bian and others

Figure 5 PSD95 enhances the expression of the Mas protein

(A) Mas protein levels were enhanced when it was co-expressed with PSD95. COS-7 cells were transfected with equal amounts of Mas wt constructs in the presence or absence of Myc–PSD95.
The Mas protein was examined by Western blotting. GAPDH was used as a loading control. (B) PSD95 overexpression enhanced Mas protein expression dependent on the association of Mas with
PSD95. COS-7 cells were transfected with FLAG–Mas wt or FLAG–Mas-V325A constructs in the presence or absence of PSD95. The Mas proteins were detected using the anti-FLAG antibody. Actin
was used as a loading control. (C) Knockdown of endogenous PSD95 expression in LN229 cells decreased Mas protein expression. LN229 cells were transfected with either PSD95 siRNA or its
scrambled control (NC). Cells were harvested 48 h after transfection, and total cellular RNA was extracted using TRIzol® reagent, or total protein was extracted with protein lysis buffer. The PSD95
mRNA levels were determined by qPCR, and the PSD95 and Mas proteins were detected using the anti-PSD95 antibody and anti-Mas antibody (* P < 0.05 compared with NC). (D) Endogenous
Mas protein expression in primary neuron cells was regulated by PSD95. Primary neuron cells were transfected with either PSD95 siRNA or its scrambled control (NC). The cells were harvested
48 h after transfection. Total cellular RNA was extracted using TRIzol® reagent, and PSD95 mRNA levels were assessed by qPCR. Total protein was extracted with protein lysis buffer, and PSD95
and Mas protein levels were detected using the anti-PSD95 antibody and anti-Mas antibody. Results in each histogram are means + − S.D. obtained by quantification of the blots from three individual
experiments (*P < 0.05 compared with NC). Molecular masses are indicated in kDa.


c The Authors Journal compilation 
c 2013 Biochemical Society
 Interaction of PSD95 and Mas receptor 353

Figure 6 PSD95 enhances the stabilization of the Mas protein

(A) Mas mRNA levels were not decreased after knocking down PSD95 expression. LN229 cells were transfected with either PSD95 siRNA or its scrambled control (NC) respectively. Mas mRNA
expression was determined by qPCR. β-Actin was used to monitor and confirm equal loading of the samples. (B) Dose-dependent Mas degradation in COS-7 cells after treatment with CHX. COS-7
cells transiently transfected with FLAG–Mas constructs were treated for 12 h with different doses of CHX, a protein synthesis inhibitor, as indicated. Mas protein expression was detected using the
anti-FLAG antibody. (C) Time course of PSD95 inhibiting the degradation of the Mas protein. COS-7 cells were transfected with FLAG–Mas wt or FLAG–Mas-V325A in the presence or absence of
Myc–PSD95. CHX (25 ng/μl) was added after 18 h of transfection. Cells were harvested at the time points indicated. The expression of Mas and PSD95 was examined by Western blotting. Results
are means + − S.D. for three independent experiments, with values within each experiment normalized to those of β-actin. Molecular masses are indicated in kDa in the gel images.


c The Authors Journal compilation 
c 2013 Biochemical Society
354 W. Bian and others

Figure 7 PSD95 modulates Mas expression through attenuation of the ubiquitin–proteasome proteolytic Mas protein
(A) The Mas protein was degraded via the proteasome pathway. C6 cells stably transfected with Mas constructs were treated with chloroquine (Chl), a lysosome inhibitor, or MG132, a proteasome
inhibitor. Mas protein expression was examined by Western blotting. (B) Mas protein levels were elevated by MG132 in a dose-dependent manner. C6 cells stably transfected with Mas were treated
with different doses of MG132 for 12 h and Mas expression was examined by Western blotting. (C) The time course of the attenuation of endogenous Mas protein degradation by MG132 in C6
cells. C6 cells were treated with MG132 (10 μM) for different times (0, 3, 9 and 12 h), endogenous Mas protein expression levels were examined by Western blotting. (D) PSD95 overexpression
reduced the ubiquitination of Mas wt. COS-7 cells were transfected with equal amounts of FLAG–Mas wt or FLAG–Mas-V325A in the presence or absence of Myc–PSD95. After 24 h of transfection,
the cells were treated with 10 μM MG132 for 12 h. Cell lysates were incubated with the anti-FLAG antibody-coupled beads to immunoprecipitate FLAG-tagged Mas. The precipitated complexes
were probed with an anti-ubiquitin (Ub) antibody (top panel) and with an anti-FLAG antibody to confirm equivalent expression of FLAG–Mas (second panel from top). Cell lysates were analysed by
Western blotting with anti-FLAG or anti-PSD95 antibodies to assess the levels of Mas (third panel from top) or PSD95 (bottom panel) respectively. The histogram shows quantification of results
(means + − S.D.) from three individual blots (*P < 0.05). mt, mutant. Molecular masses are indicated in kDa. IB, immunoblot; IP, immunoprecipitation.

pull-down combined with MS, and identified PSD95 as a novel
binding protein of Mas receptor. PSD95 (also named SAP90
and DLG4) is a PDZ domain-containing protein that possesses
three tandem PDZ domains linked to an SH3 (Src homology
3) domain and C-terminal GK (guanylate kinase) region. Mas
specifically interacted via its CT with the PDZ1-2, but not PDZ3,
domain of PSD95, which was supported by the results of GST
pull-down (Figure 2), co-immunoprecipitation (Figure 3) and
confocal co-localization (Figure 4). PSD95 PDZ1-2 domain has
been defined to interact with the consensus sequence (S/T)XV-
COOH, where X can represent any residue. However, mutation
of Mas’s valine residue at the − 1 position to alanine resulted in
complete elimination of the Mas–PSD95 interaction (Figure 2B).
This observation is not unique and has been reported previously
in other PDZ domains, such as Erbin-PDZ and ZO1-PDZ1 [26].
It is clear that other residues also contribute to specificity for a
given PDZ domain, because the (S/T)XV motif in some proteins
is insufficient for binding any of the PDZ domains of PSD95
such as the neuronal inwardly rectifying K + channels Kir3.2 and
Kir3.3 (the C-terminal sequence is ESKV in both cases) [27]
and the Na + channel Nav 1.5 (ESIV motif) [28]. This indicates
that some other residues may also affect the binding affinity of
the given PDZ domain with the (S/T)XV motif, and it is possible
that these residues or peptide sequence would not allow the single
mutation of valine to alanine at − 1 as is the case in the present
study. However, this hypothesis needs further study.
PDZ domains are modular protein-interaction domains that
play multiple roles in the regulation of the function and
signalling of GPCRs [18]. The PDZ protein MAGI-2 (membrane-
associated guanylate kinase inverted 2) associates with β 1 AR (β 1 -
adrenergic receptor) and promotes its internalization, whereas
the PDZ protein CAL reduces β 1 AR cell-surface expression and
regulates its vesicular transport [29]. Recent evidence suggests
that protein stabilization may be enhanced by binding with
some PDZ proteins. E6 protein expression in high-risk human
papillomaviruses is increased in the presence of the PDZ protein
hScrib [30]. A previous study from our group showed that the PDZ
protein CAL interacts with mGluR5a (metabotropic glutamate
receptor 5a) and enhances receptor expression by inhibiting its
ubiquitination [20].


c The Authors Journal compilation 
c 2013 Biochemical Society

In the present study, overexpression of PSD95 was shown to up-
regulate Mas protein expression (Figures 5A and 5B). Conversely,
siRNA-mediated knockdown of PSD95 significantly reduced Mas
protein levels (Figures 5C and 5D). These results indicated that
the interaction between PSD95 and Mas regulates the expression
of the Mas receptor. Our data demonstrate further that PSD95
regulation of Mas expression does not occur at the transcriptional
level, but rather by increasing Mas stabilization (Figure 6).
Intracellular protein degradation relies mainly on the lysosomal
or proteasome pathways. In the present study, we have shown that
Mas degradation was robustly retarded by MG132, a specific
proteasome inhibitor, in time- and dose-dependent manners
(Figures 7B and 7C). Proteins are targeted for degradation by the
proteasome via ubiquitination. Our results showed that PSD95
overexpression impaired the ubiquitination of Mas wt, but did
not affect the Mas-V325A mutant (Figure 7D), indicating that
the proteasome-mediated degradation of Mas is suppressed by
its association with PSD95. In a previous study, we found that
the PDZ protein NHERF1 (Na + /H + -exchanger regulatory factor
1) blocked the association of PTEN (phosphatase and tensin
homologue deleted on chromosome 10) with the ubiquitin E3
ligase NEDD4 (neural-precursor-cell-expressed developmentally
down-regulated 4), thus inhibiting the ubiquitination-dependent
PTEN degradation pathway [31]. Sequencing analysis indicates
that Mas possesses several conserved lysine-rich regions at its
CT (the last 40 amino acids of Mas-CT are S286 KKKRFKESL-
KVVLTRAFKDEMQPRRQKDNCNTVTVETVV325 ), and Mas
Lys313 is most likely to be one of the Mas protein ubiquitination
sites by predication using UbPred (http://www.ubpred.org). It
is possible that the PSD95–Mas interaction may also interfere
with the binding of an ubiquitin E3 ligase to Mas-CT, resulting
in the inhibition of Mas ubiquitination and suppression of receptor
degradation. However, this possibility needs to be investigated
further.
The Mas protein is expressed in different brain areas including
the hippocampus, forebrain and piriform cortex [32]. PSD95 is
also expressed in the prefrontal cortex and hippocampus of the
brain [33,34]. Together with our current results, this suggests
that Mas and PSD95 may physically interact in the brain, where
the association could have a potentially important role. The
Mas receptor in the brain is known to be involved in neural
plasticity, memory, anxiety and LTP associated with PI3K/Akt-
mediated nitric oxide release and c-Src activation [12,35]. Indeed,
by binding nNOS (neuronal nitric oxide synthase) or Src PTKs
(protein tyrosine kinases), PSD95 can also regulate the learning,
memory and neural plasticity and mediate LTP in synapses
of the hippocampus [36–38]. Because PSD95 can associate
directly with Mas and nNOS respectively, it may function as a
scaffold protein to mediate the formation of a Mas–PSD95–nNOS
macromolecular complex involved in the regulation of Ang-(1–
7)–Mas signalling and its neurobiological functions. However,
this possibility needs to be explored further.
In the brain, hypothalamic Ang-(1–7) plays an important role
in blood pressure regulation via its receptor Mas [39]. The SHR
(spontaneously hypertensive rat) exhibits a chronically stimulated
brain angiotensin system and cerebrovascular hypertrophy [40].
Saavedra and colleagues explored species differences in gene
expression in 12-week-old SHR male rats and WKY (Wistar
Kyoto) rats using the Affymetrix Rat Genome U34 Array (GEO
ID: GDS2070), and found no statistically significant differences in
Mas mRNA expression in brain microvascular tissue (P = 0.17),
whereas PSD95 mRNA expression was significantly reduced by
approximately 47 % in SHR compared with WKY male rats (P =
0.02) [40]. The present results, however, showed that PSD95
down-regulation destabilized the Mas protein resulting in low
Interaction of PSD95 and Mas receptor 355
Mas receptor levels (Figure 5). Thus one possible molecular
mechanism of hypertension in SHR is that the low level of
PSD95 mRNA expression may weaken the anti-hypertension
effect of Mas as a result of Mas receptor destabilization and
down-regulation. Further studies are necessary to confirm this
hypothesis.
In summary, the PDZ protein PSD95 was identified as a
novel Mas receptor-interacting molecule. The C-terminus of
Mas specifically interacts with the PDZ1-2 domain of PSD95.
Functional studies showed that Mas protein expression was up-
regulated through its interaction with PSD95, which suppressed
Mas ubiquitination and proteasomal degradation. Therefore our
findings reveal a new mechanism for the regulation of Mas protein
expression by PSD95 involving the inhibition of ubiquitination-
dependent degradation. The elucidation of this mechanism of
PSD95-mediated regulation may help to clarify further the
functions of the Mas receptor in the central nervous system.
AUTHOR CONTRIBUTION
Junqi He and Songlin Wang designed and directed experiments. Weihua Bian and Licui
Sun carried out the most experiments and data analysis. Longyan Yang performed the
additional experiments for the revised paper. Jifeng Li performed the endogenous co-
immunoprecipitation experiments. Jia Hu performed the proteomic experiments. Shuai
Zheng and Ruihan Guo performed protein stabilization assays. Duiping Feng, Qian Ma,
Xiaocui Shi and Jianguo Xu contributed to database searches. Ying Xiong, Ran Song and
Xiaomei Yang contributed to data analysis. Weihua Bian, Licui Sun and Junqi He co-wrote
the paper. All authors contributed to and have approved the final paper.
FUNDING
This work was supported by the National Natural Science Foundation of the People’s
Republic of China [grant numbers 81272887 and 81141033], Beijing Natural Science
Foundation [grant numbers 7131003 and 5112007] and the Foundation of Beijing
Educational Committee [grant numbers KZ201010025026 and SQKM201210025002].
REFERENCES
1 Santos, R. A., Ferreira, A. J. and Simoes, E. S. A. C. (2008) Recent advances in the
angiotensin-converting enzyme 2-angiotensin(1–7)-Mas axis. Exp. Physiol. 93, 519–527
2 Iusuf, D., Henning, R. H., van Gilst, W. H. and Roks, A. J. (2008) Angiotensin-(1–7):
pharmacological properties and pharmacotherapeutic perspectives. Eur. J. Pharmacol.
585, 303–312
3 Kostenis, E., Milligan, G., Christopoulos, A., Sanchez-Ferrer, C. F., Heringer-Walther, S.,
Sexton, P. M., Gembardt, F., Kellett, E., Martini, L., Vanderheyden, P. et al. (2005)
G-protein-coupled receptor Mas is a physiological antagonist of the angiotensin II type 1
receptor. Circulation 111, 1806–1813
4 Lara, L. S., Vives, D., Correa, J. S., Cardozo, F. P., Marques-Fernades, M. F., Lopes, A. G.
and Caruso-Neves, C. (2010) PKA-mediated effect of MAS receptor in counteracting
angiotensin II-stimulated renal Na + -ATPase. Arch. Biochem. Biophys. 496, 117–122
5 Xu, P., Sriramula, S. and Lazartigues, E. (2011) ACE2/ANG-(1–7)/Mas pathway in the
brain: the axis of good. Am. J. Physiol. Regul. Integr. Comp. Physiol. 300, R804–R817
6 Martin, K. A. and Hockfield, S. (1993) Expression of the mas proto-oncogene in the rat
hippocampal formation is regulated by neuronal activity. Mol. Brain Res. 19, 303–309
7 Hellner, K., Walther, T., Schubert, M. and Albrecht, D. (2005) Angiotensin-(1–7) enhances
LTP in the hippocampus through the G-protein-coupled receptor Mas. Mol. Cell.
Neurosci. 29, 427–435
8 Walther, T., Voigt, J. P., Fink, H. and Bader, M. (2000) Sex specific behavioural alterations
in Mas-deficient mice. Behav. Brain Res. 107, 105–109
9 Hocht, C., Gironacci, M. M., Mayer, M. A., Schuman, M., Bertera, F. M. and Taira, C. A.
(2008) Involvement of angiotensin-(1–7) in the hypothalamic hypotensive effect of
captopril in sinoaortic denervated rats. Regul. Pept. 146, 58–66
10 Santos, R. A., Castro, C. H., Gava, E., Pinheiro, S. V., Almeida, A. P., Paula, R. D., Cruz,
J. S., Ramos, A. S., Rosa, K. T., Irigoyen, M. C. et al. (2006) Impairment of in vitro and
in vivo heart function in angiotensin-(1–7) receptor MAS knockout mice. Hypertension
47, 996–1002
11 Rabelo, L. A., Alenina, N. and Bader, M. (2011) ACE2–angiotensin-(1–7)–Mas axis and
oxidative stress in cardiovascular disease. Hypertens. Res. 34, 154–160

c The Authors Journal compilation 
c 2013 Biochemical Society
356 W. Bian and others
12 Sampaio, W. O., Souza dos Santos, R. A., Faria-Silva, R., da Mata Machado, L. T.,
Schiffrin, E. L. and Touyz, R. M. (2007) Angiotensin-(1–7) through receptor Mas mediates
endothelial nitric oxide synthase activation via Akt-dependent pathways. Hypertension
49, 185–192
13 Rabelo, L. A., Xu, P., Todiras, M., Sampaio, W. O., Buttgereit, J., Bader, M., Santos, R. A.
and Alenina, N. (2008) Ablation of angiotensin (1–7) receptor Mas in C57Bl/6 mice
causes endothelial dysfunction. J. Am. Soc. Hypertens. 2, 418–424
14 Nie, W., Yan, H., Li, S., Zhang, Y., Yu, F., Zhu, W., Fan, F. and Zhu, J. (2009)
Angiotensin-(1–7) enhances angiotensin II induced phosphorylation of ERK1/2 in mouse
bone marrow-derived dendritic cells. Mol. Immunol. 46, 355–361
15 Zhang, F., Hu, Y., Xu, Q. and Ye, S. (2010) Different effects of angiotensin II and
angiotensin-(1–7) on vascular smooth muscle cell proliferation and migration. PLoS ONE
5, e12323
16 Young, D., Waitches, G., Birchmeier, C., Fasano, O. and Wigler, M. (1986) Isolation and
characterization of a new cellular oncogene encoding a protein with multiple potential
transmembrane domains. Cell 45, 711–719
17 Daaka, Y. (2012) S-nitrosylation-regulated GPCR signaling. Biochim. Biophys. Acta
1820, 743–751
18 Romero, G., von Zastrow, M. and Friedman, P. A. (2011) Role of PDZ proteins in
regulating trafficking, signaling, and function of GPCRs: means, motif, and opportunity.
Adv. Pharmacol. 62, 279–314
19 Shao, X., Zhu, L., Wang, Y., Lu, Y., Wang, W., Zhu, J., Shen, Y., Xia, J. and Luo, J. (2010)
Threonine 82 at the PDZ domain of PICK1 is critical for AMPA receptor interaction and
localization. Neurochem. Int. 56, 962–970
20 Cheng, S., Zhang, J., Zhu, P., Ma, Y., Xiong, Y., Sun, L., Xu, J., Zhang, H. and He, J.
(2010) The PDZ domain protein CAL interacts with mGluR5a and modulates receptor
expression. J. Neurochem. 112, 588–598
21 Chen, V. C., Li, X., Perreault, H. and Nagy, J. I. (2006) Interaction of zonula occludens-1
(ZO-1) with α-actinin-4: application of functional proteomics for identification of PDZ
domain-associated proteins. J. Proteome Res. 5, 2123–2134
22 He, J., Bellini, M., Inuzuka, H., Xu, J., Xiong, Y., Yang, X., Castleberry, A. M. and Hall,
R. A. (2006) Proteomic analysis of β 1 -adrenergic receptor interactions with PDZ scaffold
proteins. J. Biol. Chem. 281, 2820–2827
23 Li, C., Zhou, Y., Liu, Z., Tuo, J., Hu, N. and Guan, H. (2012) Spatiotemporal expression of
postsynaptic density 95 in rat retina after optic nerve injury. J. Mol. Neurosci. 46,
595–605
24 Chaudhury, A., He, X. D. and Goyal, R. K. (2009) Role of PSD95 in membrane association
and catalytic activity of nNOSα in nitrergic varicosities in mice gut. Am. J. Physiol.
Gastrointest. Liver Physiol. 297, G806–G813
25 Gironacci, M. M., Adamo, H. P., Corradi, G., Santos, R. A., Ortiz, P. and Carretero, O. A.
(2011) Angiotensin (1–7) induces MAS receptor internalization. Hypertension 58,
176–181
26 Zhang, Y., Yeh, S., Appleton, B. A., Held, H. A., Kausalya, P. J., Phua, D. C., Wong, W. L.,
Lasky, L. A., Wiesmann, C., Hunziker, W. and Sidhu, S. S. (2006) Convergent and
divergent ligand specificity among PDZ domains of the LAP and zonula occludens (ZO)
families. J. Biol. Chem. 281, 22299–22311
Received 19 December 2012/24 May 2013; accepted 24 May 2013
Published as BJ Immediate Publication 24 May 2013, doi:10.1042/BJ20121885

c The Authors Journal compilation 
c 2013 Biochemical Society
27 Nehring, R. B., Wischmeyer, E., Doring, F., Veh, R. W., Sheng, M. and Karschin, A. (2000)
Neuronal inwardly rectifying K + channels differentially couple to PDZ proteins of the
PSD-95/SAP90 family. J. Neurosci. 20, 156–162
28 Gee, S. H., Quenneville, S., Lombardo, C. R. and Chabot, J. (2000) Single-amino acid
substitutions alter the specificity and affinity of PDZ domains for their ligands.
Biochemistry 39, 14638–14646
29 He, J., Bellini, M., Xu, J., Castleberry, A. M. and Hall, R. A. (2004) Interaction with cystic
fibrosis transmembrane conductance regulator-associated ligand (CAL) inhibits
β 1 -adrenergic receptor surface expression. J. Biol. Chem. 279, 50190–50196
30 Nicolaides, L., Davy, C., Raj, K., Kranjec, C., Banks, L. and Doorbar, J. (2011)
Stabilization of HPV16 E6 protein by PDZ proteins, and potential implications for genome
maintenance. Virology 414, 137–145
31 Yang, L., Wang, Y., Chen, P., Hu, J., Xiong, Y., Feng, D., Liu, H., Zhang, H., Yang, H. and
He, J. (2011) Na + /H + exchanger regulatory factor 1 (NHERF1) is required for the
estradiol-dependent increase of phosphatase and tensin homolog (PTEN) protein
expression. Endocrinology 152, 4537–4549
32 Metzger, R., Bader, M., Ludwig, T., Berberich, C., Bunnemann, B. and Ganten, D. (1995)
Expression of the mouse and rat mas proto-oncogene in the brain and peripheral tissues.
FEBS Lett. 357, 27–32
33 Ohnuma, T., Kato, H., Arai, H., Faull, R. L., McKenna, P. J. and Emson, P. C. (2000) Gene
expression of PSD95 in prefrontal cortex and hippocampus in schizophrenia.
NeuroReport 11, 3133–3137
34 Du, Y., Li, C., Hu, W. W., Song, Y. J. and Zhang, G. Y. (2009) Neuroprotection of
preconditioning against ischemic brain injury in rat hippocampus through inhibition of
the assembly of GluR6–PSD95–mixed lineage kinase 3 signaling module via nuclear and
non-nuclear pathways. Neuroscience 161, 370–380
35 Lazaroni, T. L., Raslan, A. C., Fontes, W. R., de Oliveira, M. L., Bader, M., Alenina, N.,
Moraes, M. F., Dos Santos, R. A. and Pereira, G. S. (2012) Angiotensin-(1–7)/Mas axis
integrity is required for the expression of object recognition memory. Neurobiol. Learn.
Mem. 97, 113–123
36 Sakagami, H., Sanda, M., Fukaya, M., Miyazaki, T., Sukegawa, J., Yanagisawa, T., Suzuki,
T., Fukunaga, K., Watanabe, M. and Kondo, H. (2008) IQ-ArfGEF/BRAG1 is a guanine
nucleotide exchange factor for Arf6 that interacts with PSD-95 at postsynaptic density of
excitatory synapses. Neurosci. Res. 60, 199–212
37 Ehrlich, I., Klein, M., Rumpel, S. and Malinow, R. (2007) PSD-95 is required for
activity-driven synapse stabilization. Proc. Natl. Acad. Sci. U.S.A. 104, 4176–4181
38 Xu, J., Liu, Z. A., Pei, D. S. and Xu, T. J. (2010) Calcium/calmodulin-dependent kinase II
facilitated GluR6 subunit serine phosphorylation through GluR6–PSD95–CaMKII
signaling module assembly in cerebral ischemia injury. Brain Res. 1366,
197–203
39 Gironacci, M. M., Brosnihan, K. B., Ferrario, C. M., Gorzalczany, S., Verrilli, M. A.,
Pascual, M., Taira, C. and Pena, C. (2007) Increased hypothalamic angiotensin-(1–7)
levels in rats with aortic coarctation-induced hypertension. Peptides 28, 1580–1585
40 Ando, H., Zhou, J., Macova, M., Imboden, H. and Saavedra, J. M. (2004) Angiotensin II
AT1 receptor blockade reverses pathological hypertrophy and inflammation in brain
microvessels of spontaneously hypertensive rats. Stroke 35, 1726–1731